This revised application responds to PA-10-069. Alcoholics Anonymous (AA) is the most popular self-help program for persons with alcohol-related problems, and 12-step therapy, based on AA doctrine and practice, is the prevailing alcohol treatment model in the United States. Understandably, then, hundreds of investigations on community-based AA have been conducted, and recent narrative reviews of this empirical literature are relatively common. Narrative reviews are of limited value as they tend to review only "statistically significant" findings, they do not provide estimates of the magnitude of effects and, by not providing such estimates, these reviews miss the opportunity to alert researchers to the most fertile areas for future investigations. To overcome these limitations we propose to do a comprehensive review of the empirical AA literature (1993- 2009) using advanced meta-analytic techniques. It has been 19 years since the last comprehensive review of this kind (Emrick et al., 1993), and in that time about 450-500 AA peer- reviewed papers have been published. This review has three aims: (1) we will summarize the AA literature from 1993-2009 focusing on AA affiliation profiles and the magnitude and nature of AA- related benefit. We will do this with much improved procedural and statistical approaches relative to our 1993 and 1996 meta-analytic reviews, (2) we will review three new areas of AA research, a meta-analytic synthesis made possible because of the surge of AA research since 1993. The areas are: (a) mechanisms of change in AA, (b) special populations and AA, and (c) long-term AA affiliation and outcome. Summaries of these new areas will provide practitioners and researchers with a unique perspective in understanding how, and for whom, AA may be helpful. Related, we plan to identify, code, and analyze non-AA self-help peer-reviewed papers. (3) We will disseminate meta-analytic papers to high impact journals, make available the data set to investigators, and distribute, free of charge, a CD that contains a PDF file with the references for all qualitative and quantitative papers identified in the meta-analytic search, with accompanying PDF reproduced articles when copyright laws permit. PUBLIC HEALTH RELEVANCE: This project will apply advanced meta-analytic techniques to review and summarize the AA-based research literature from 1993 to 2009. We will review English and Spanish peer-reviewed papers, and we will also include non-12 step self-help papers. In addition to identifying AA affiliation profiles and the nature and magnitude of AA-related benefit, this comprehensive review will synthesize findings in three new areas of AA research that have proliferated since 1993: (1) mechanisms of change in AA, (2) special populations and AA, and (3) long-term AA careers and outcome. Findings will provide both: (1) evidence-based recommendations to clinicians about who does and does not benefit from AA, and (2) clear recommendations to researchers about the most fertile areas for future mutual-help investigations.